Abstracts

Rationale: Electroencephalogram (EEG) is being increasingly utilized for evaluation and diagnosis of a multitude of conditions beyond epilepsy, such as altered mental status or confusion, by both neurologists and non-neurologists. However, yield of EEG in finding epileptiform discharges in patients presenting with altered mental status is unknown. In this retrospective study, we investigate the diagnostic utility of EEG in this group of patients and the factors that may impact its yield. Methods: We reviewed the medical records of 55 randomly selected patients 18 years of age or older admitted to the Upstate University Hospital Downtown or Community Campus between July 1, 2010 to November 30, 2018 who had completed an inpatient EEG for evaluation of acute altered mental status or confusion. Demographic data, medical and social history as well as information about the ordering service (Neurology versus Non-Neurology) was collected. The EEG findings, based on the report, were categorized into two sub-groups: Group A with normal, or abnormal without epileptiform findings; and Group B with abnormal epileptiform findings. Results: The EEG findings of 46 patients (83.6%) were categorized as Group A, and 9 patients (16.4%) as Group B. There was no statistical difference between these 2 respective groups in age (66.4 +/- 11.9 years vs. 65.7 +/- 12.5 years, p=0.87), gender (male: 54.3% [n=25] vs. 44.4% [n=4], p=0.59), previous history of seizures (23.9% [n=11] vs n=11.1% [n=1], p=0.67) or previous history of encephalopathy (37% [n=17] vs 11.1% [n=1], p=0.24). However, when there was a neurologist evaluation (49 out of the 55 records) and there was a clinical suspicion of seizure etiology by the neurologist prior to the completion of EEG, 28.6% of the EEG results fell into Group B, and when there was not such a clinical suspicion by neurologist only 7.1% showed epileptiform abnormalities (p=0.04). This difference did not reach statistical significance when the clinical suspicion of etiology was explicitly documented by a non-neurologist evaluation (42 out of the 55 records): the clinical suspicion of seizure etiology by a non-neurologist was associated with finding of epileptiform discharges (Group B) in 17.4% of patients versus in 5.3% of patients with no clinical suspicion (p=0.23). Conclusions: EEG yield in finding epileptiform abnormalities in patients presenting with altered mental status is higher when there is a clinical suspicion for underlying seizure etiology by a neurologist evaluation, however, this trend does not clearly reflect in the clinical evaluations by a non-neurologist. These findings are significant because not only do they help gain insight into the diagnostic utility and yield of EEG studies in patients presenting with encephalopathy but also illustrate the significance of a neurologically focused history and clinical evaluation in these patients. Funding: No funding